GitHub Copilot CLI vs OpenCode: pricing, features and which to choose

GitHub Copilot CLI is GitHub's terminal-native coding agent, launched in public preview in September 2025. It is distributed as the @github/copilot npm package and uses a GitHub account for authentication. The product works with local repositories while connecting to GitHub context, including repository history, issues, and pull requests. GitHub documents support for Linux, macOS, and Windows, with PowerShell 6 or later required on Windows. The current CLI is distinct from the older gh-copilot extension. GitHub announced that gh-copilot would stop working on October 25, 2025, and presented the agentic CLI as its replacement. That distinction matters because the older extension offered command suggestions, while the current product can plan and execute multi-step repository work.

A normal session starts with copilot in a repository and a natural-language request. The agent can inspect files, propose a plan, edit code, debug, refactor, and run terminal commands. GitHub context lets it work from repository tasks, issues, and pull requests, while the included GitHub MCP server connects the session to GitHub capabilities. Custom MCP servers and custom agents extend that setup. GitHub says each file change and command execution requires explicit approval. Release notes also describe streaming output, parallel tool calls, model selection, image input, session continuation, delegated work, and granular tool permissions. The workflow combines shell access with repository context, but users must review proposed edits and commands before allowing them, particularly when a request can affect files or external GitHub state.

Copilot CLI suits developers who keep repositories, issues, and pull requests in GitHub and want organization controls around terminal agents. Its limits are clear: it requires GitHub authentication, access depends on a Copilot plan, and agent work consumes credits. The product also depends on remote GitHub-hosted AI services, so it is not a local-only runtime or a self-hosted deployment. Crush offers provider and local-model choice, Claude Code centers Anthropic's service, and OpenCode emphasizes support for multiple providers. Pick Copilot CLI when GitHub context and centralized policy outweigh endpoint freedom. Pick Crush for local models, Claude Code for Anthropic-centered workflows, or OpenCode for provider flexibility.

OpenCode is an open source coding agent for terminal, desktop, and IDE workflows. The public site reports more than 215,000 GitHub stars, 950 contributors, 20,000 commits, and more than 16 million developers monthly, figures the company can change. The repository identifies the software as MIT licensed. Current documentation distinguishes OpenCode 1 and OpenCode 2, which share the same command name and cannot coexist under default installation paths. Version 2.1.2 is dated October 21, 2026. OpenCode is best understood as a client and agent layer rather than one hosted assistant. It can inspect repositories, explain code, edit files, run commands, and keep multiple sessions. The MIT license permits commercial modification and distribution subject to its conditions.

A typical workflow starts in a project directory with the opencode command. The documentation describes an /init command that creates an AGENTS.md file for project structure and coding conventions. OpenCode separates a build agent from a plan agent: the build agent can edit files and run commands, while the plan agent is read-only by default and asks before bash commands. Sessions can use LSP integration, provider selection, and a general subagent. Developers can connect OpenAI, Anthropic, Google, xAI, GitHub Copilot, ChatGPT Plus or Pro, local models, OpenCode Zen, or OpenCode Go. The terminal, desktop, and IDE clients support the same provider choice. This workflow is adaptable, although the account, model, quota, and data terms depend on the selected route.

OpenCode's main tradeoff is choice. Provider selection can reduce lock-in, but it also makes billing, privacy, and troubleshooting less predictable than a single-vendor product. The Go privacy table says most listed models do not use prompts for training, while Muse Spark models permit training and Gemini 4 Argon Contributor may use prompts and completions for future Google models. Compared with Claude Code, OpenCode offers local models and a broader provider route. Compared with Cursor, it is more terminal centered and MIT licensed. Compared with Aider, it adds desktop and IDE clients. Pick OpenCode for repository work when provider flexibility matters more than one unified account and bill.